Etherman's Indo-Uralic Thread

[Mbwa]

So... Ural-Altaic's not a crazy idea, Indo-Uralic's not either... I think it's time to start reconstructing Proto-World.

Ha, kidding. More seriously, I'll make a little chart of some common words in a bit. I'll probably do a bit of this for the Ural-Altaic theory too, not because I think I'm a linguist, but because I have a lot of free time and I think it's cool.

[Octaviano]

Unfortunately, some of the these PIE-PU correspondences might reflect cross-borrowings (usually from IE into PU) and even (substrate) loanwords from a third party.

That's always a danger. That's why it's important to establish regular correspondences (something you've been loathe to do).

I'm affraid that sound correspondences alone won't tell you an inherited word from a borrowing.

For example, IE *salHk(^)- 'willow' (a regional NW word) looks like a substrate loanword (even Matasovic recognizes this!) connected to Tungus-Manchu *dʒali-kta 'hawthorn, willow'

Alternatively this could be a Nostratic root reflected in PIE, PU, and PA.

Unfortunately, this isn't a PIE word.

This would undermine any Indo-Uralic theory.

Given the regular correspondences in lexical roots, grammatical words, and morphology a few borrowings won't hurt anything.

What if these "few" borrowings would turn to be "many" ones?

[Radagast]

Unfortunately, some of the these PIE-PU correspondences might reflect cross-borrowings (usually from IE into PU) and even (substrate) loanwords from a third party.

That's always a danger. That's why it's important to establish regular correspondences (something you've been loathe to do).

I'm affraid that sound correspondences alone won't tell you an inherited word from a borrowing.

And how is this a reason not to establish sound correspondences? In order to suggest a borrowing you need either a likely donor etymon or to show that the word is phonologically alien - if you can do both that strengthens the case. If you do neither then you are not doing linguistics at all but just flinging out unsupported claims and speculation.

[Octaviano]

I'm affraid that sound correspondences alone won't tell you an inherited word from a borrowing.

And how is this a reason not to establish sound correspondences? In order to suggest a borrowing you need either a likely donor etymon or to show that the word is phonologically alien - if you can do both that strengthens the case. If you do neither then you are not doing linguistics at all but just flinging out unsupported claims and speculation.

In the first place, I should made clear this kind of "problem" is intrinsecal to every traditional "Nostratic" theory (and I-U happens to be one of them).

The relevants questions here are:
1) How can we tell an inherited PIE word from a borrowing?
2) How can we tell an inherited PU word from a borrowing (specially those from IE)?

For example, I've just shown why we can't consider *salHk(^)- 'willow' to be an inherited PIE word. Although recovering the PIE inherited lexicon at the Mesolithic stage mightn't be an straightforward task, it's something it MUST be done prior to building any "Nostratic" theory.

[Dewrad]

For example, I've just shown why we can't consider *salHk(^)- 'willow' to be an inherited PIE word.

No you haven't:

Unfortunately, this isn't a PIE word.

You've just made an unsupported statement. You haven't shown anything.

[WeepingElf]

I'm affraid that sound correspondences alone won't tell you an inherited word from a borrowing.

Oh, they do. Borrowings show irregular correspondences because they did not undergo the same sound changes as inherited words. (No two languages have the same phonological history; hence, a word that was at some point borrowed from a related language differs in its phonological development from a native word.) For example, English paternal can be shown to be a borrowing because it hasn't undergone the *p > f shift inherited words did undergo.

[roninbodhisattva]

I'm affraid that sound correspondences alone won't tell you an inherited word from a borrowing.

So do you have an actual methodology for distinguishing loans from a borrowing?

[David McCann]

Well we have a Vasco-Caucasian thread, and a Ural-Altaic thread, so why not an Indo-Uralic thread? The Indo-Uralic hypothesis (with or without the wider Nostratic hypothesis) is nothing new (been around for decades) but not accepted by mainstream linguists. However, it's not a crackpot idea either. It's been/is supported by such mainstream linguists as Kortlandt and Pederson.

You said: 1) it's not accepted by mainstream linguists and 2) it's been/is supported by such mainstream linguists as Kortlandt and Pederson.

A nice contradiction!!!

There tends to be a Europe/America split. The relationship has been widely accepted in Europe for a century, but raises inexplicable passions in the USA.

[Octaviano]

For example, I've just shown why we can't consider *salHk(^)- 'willow' to be an inherited PIE word.

No you haven't:

Unfortunately, this isn't a PIE word.

You've just made an unsupported statement. You haven't shown anything.

AFAIK, this word is only attested in Celtic, Germanic and Latin (probably a Gaulish loanword), so it's most probably a substrate loanword, not a native word.

[Etherman]

AFAIK, this word is only attested in Celtic, Germanic and Latin (probably a Gaulish loanword), so it's most probably a substrate loanword, not a native word.

When was any Tungus-Manchu language a substrate in Western Europe?

[Octaviano]

Borrowings show irregular correspondences because they did not undergo the same sound changes as inherited words. (No two languages have the same phonological history; hence, a word that was at some point borrowed from a related language differs in its phonological development from a native word.)

It depends on the antiquity of the borrowing. For example, Neolithic loanwords in IE like *pork^o- 'piglet' are phonetically indistinguishable from native PIE words.

[Octaviano]

AFAIK, this word is only attested in Celtic, Germanic and Latin (probably a Gaulish loanword), so it's most probably a substrate loanword, not a native word.

When was any Tungus-Manchu language a substrate in Western Europe?

There're some traces of an Altaic substrate in NW Europe.

[Radagast]

AFAIK, this word is only attested in Celtic, Germanic and Latin (probably a Gaulish loanword), so it's most probably a substrate loanword, not a native word.

When was any Tungus-Manchu language a substrate in Western Europe?

There're some traces of an Altaic substrate in NW Europe.

Really, how interesting. Tell me more about that please?

[TomHChappell]

The higher the latitude, the shorter the distance across Eurasia. Assuming an entirely northerly route, Kolyma, Yenisei, Murmansk, and Tromso, aren't all that far apart -- are they?

[dhok]

I'm affraid that sound correspondences alone won't tell you an inherited word from a borrowing.

Oh, they do. Borrowings show irregular correspondences because they did not undergo the same sound changes as inherited words. (No two languages have the same phonological history; hence, a word that was at some point borrowed from a related language differs in its phonological development from a native word.) For example, English paternal can be shown to be a borrowing because it hasn't undergone the *p > f shift inherited words did undergo.

I'll play devil's advocate here.

When we are talking about something as long ago as Octaviano is putting forward, you're likely to miss a sound change or two.

To take an example, English's p>f shift was part of a chain shift, and a few p's did appear, from PIE *b. Now, we do know this is a borrowing, because we have detailed documentation of the parent language and the transfer of the word, and the semantics don't work out any other way. (Plus, PIE *b was incredibly rare.)

Now, here's the problem. It is entirely possible that borrowings go through sound changes involving mergers, splitters and chain shifts-in fact, it's probable- and none of the languages whence these borrowings came are documented first-hand at all, nor are they particularly well-reconstructed, with the exception of Uralic. Add in sporadic changes, and you're really screwed. So, we wouldn't necessarily be able to tell.

[gsandi]

Borrowings show irregular correspondences because they did not undergo the same sound changes as inherited words. (No two languages have the same phonological history; hence, a word that was at some point borrowed from a related language differs in its phonological development from a native word.)

It depends on the antiquity of the borrowing. For example, Neolithic loanwords in IE like *pork^o- 'piglet' are phonetically indistinguishable from native PIE words.

As is often the case, I don't know whether you are making a statement (= I know that this is the case) or you are making a hypothesis (= I propose that this might have been the case, and if it was, it sure explains some things).

If it's the first, you may believe this to be the truth, but I know of no evidence that would confirm it. How do you know that it's a loanword, and even if so, how do you know that it was a loanword in neolithic times? What were neolithic times anyway? Parts of Europe were still in the neolithic when others were in the chalcolithic (copper age) and the bronze age. And when much of Europe was in the neolithic, parts were still living a hunter-gathering existence, piling up enormous mounds of shells along the Baltic shores for example.

If it's the second, you should be clearer. At least make a disclaimer somewhere (as I tend to do) that you are advancing a hypothesis, because you think it may explain some developments better than mainstream (or simply other) explanations.

[Octaviano]

Borrowings show irregular correspondences because they did not undergo the same sound changes as inherited words. (No two languages have the same phonological history; hence, a word that was at some point borrowed from a related language differs in its phonological development from a native word.)

It depends on the antiquity of the borrowing. For example, Neolithic loanwords in IE like *pork^o- 'piglet' are phonetically indistinguishable from native PIE words.

As is often the case, I don't know whether you are making a statement (= I know that this is the case) or you are making a hypothesis (= I propose that this might have been the case, and if it was, it sure explains some things).

If it's the first, you may believe this to be the truth, but I know of no evidence that would confirm it. How do you know that it's a loanword, and even if so, how do you know that it was a loanword in neolithic times?

This root is attested in PNC *wHa:rtɬɬ’wə 'boar, pig' and Starostin included it in his list of loanwords into PIE (although I independently reached the same conclusion). In my hypothesis, this word was borrowed by the Neolithic ancestor of traditional (Late) PIE from a Vasco-Caucasian language spoken by farmers, probably in the Balkans.

AFAIK, the native PIE (that is, inherited from its Mesolithic ancestor) word for 'pig' is *suH-.

[gsandi]

Borrowings show irregular correspondences because they did not undergo the same sound changes as inherited words. (No two languages have the same phonological history; hence, a word that was at some point borrowed from a related language differs in its phonological development from a native word.)

It depends on the antiquity of the borrowing. For example, Neolithic loanwords in IE like *pork^o- 'piglet' are phonetically indistinguishable from native PIE words.

As is often the case, I don't know whether you are making a statement (= I know that this is the case) or you are making a hypothesis (= I propose that this might have been the case, and if it was, it sure explains some things).

If it's the first, you may believe this to be the truth, but I know of no evidence that would confirm it. How do you know that it's a loanword, and even if so, how do you know that it was a loanword in neolithic times?

This root is attested in PNC *wHa:rtɬɬ’wə 'boar, pig' and Starostin included it in his list of loanwords into PIE (although I independently reached the same conclusion). In my hypothesis, this word was borrowed by the Neolithic ancestor of traditional (Late) PIE from a Vasco-Caucasian language spoken by farmers, probably in the Balkans.

AFAIK, the native PIE (that is, inherited from its Mesolithic ancestor) word for 'pig' is *suH-.

OK, so this is according to your hypothesis.

I don't feel competent to comment on PNC reconstructions. On the other hand, given that PIE had a *w, in fact (in my hypothesis, anyway) in laryngeal times even a *x_w, I don't see why *wHa:rtɬɬ’wə wouldn't have been borrowed as, say PIE *wa:rtwa:, *x_wa:rtwa: or something along those lines. And how do we know that the borrowing wasn't in the other direction?

The fact that there was a PIE word *suH- for 'pig' doesn't preclude the existence of other words meaning pig, sow, boar, piglet, wild boar etc.

As for what was inherited from the Mesolithic, I wouldn't dare to make such predictions. What on earth would I base such a conclusion on? We know nothing about the Mesolithic languages of Eurasia.

[Salmoneus]

We don't; Octaviano does.

[Etherman]

Some Indo-Uralic morphological correspondences:

1) PIE accusative *-m, PU accusative *-m

2) PIE ablative *-d (possibly also neuter nom/acc in neuter pronouns), PU partive/separative *-t(V)

3) PIE nominalizer *-me/mo, PU resultative or passive participle *-ma

4) PIE denomial and deverbative adjectival suffix *-lo, PU deverbative and denomial suffix *-la

5) PIE nominal/adjectival suffix *-ko (and/or *-go), PU deverbal/denomial suffix *-ka

6) PIE past passive participle *-no, PU nominal/adjectival suffix *-n(V)

7) PIE participle *-to, PU infinitive and participle *t

8. PIE participle *-nt, PU participle *-nt

9) PIE imperative *-k(V), PU imperative *-kV

10) PIE plural *-j/-e-, PU plural *-j

11) PIE dual *-h1, PU dual *-k

12) PIE plural *-s, PU plural *-t (this has been suggested by others but I'm personally skeptical)

13) PIE *-h2e 1st person perfect verbal suffix, PU *-k 1st person verbal suffix

14) PIE superlative *-mh1, PU comparative *-mpa

15) PIE sigmatic aorist *-s, PU past *-s'a

16) PIE "and" clitic *kwe, PU "and" suffix *-ka

17) PIE negative particle *n(V), PU negative particle *ne

[Basilius]

On the other hand, given that PIE had a *w, in fact (in my hypothesis, anyway) in laryngeal times even a *x_w, I don't see why *wHa:rtɬɬ’wə wouldn't have been borrowed as, say PIE *wa:rtwa:, *x_wa:rtwa: or something along those lines.

Starostin viewed this word as part of a set of probable loans for which he proposed a set of regular sound correspondences.

According to those, a PNC lateral obstruent must be reflected as a velar in PIE (as suggested by 5 examples including the one being discussed).

The initial bilabial stop isn't surprising either, since in fact all the EC branches except Proto-Lezgic have b- (the regular reflex of PNC *w(H)-) in this word. The default reflex of both PNC w(H)- and PNC b- in PIE, according to Starostin, would be bh-; with the etymon in question, this would violate PIE phonotactics, which explains the actual PIE p-. (There are other instances of voiced aspirates replaced with voiceless stops in similar environments in the same set. Also, Starostin points to PGerm *barha- 'porcus castratus' as potentially reflecting the more archaic shape of the same stem in PIE.)

And how do we know that the borrowing wasn't in the other direction?

The other direction is not probable since (1) PNC is older than PIE and (2) if the word was borrowed into individual NC langs, it is difficult to explain reflexes with initial w- (in Lesgic), as well as the lateral in Tsezic and NW Caucasian.

Also, the whole set of potential loans analyzed by Starostin shows that many PNC contrasts (including those retained in most NC branches, e. g. velar :: uvular) are neutralized, and the words seem to have been adjusted to PIE phonotactics (e. g. by simplifying the clusters not tolerated in PIE, besides altering the modal features of stops like in *pork^o-).

Actually, Starostin argues that the whole set of rather consistent sound correspondences points to a single donor language, more specifically, a descendant of PNC (chronologically older than "broad" PIE) characterized by:

(a) PNC *w- -> b word-initially;
(b) PNC *l and *r dropped in certain types of clusters;
(c) PNC *l -> r intervocalically and before consonants;
(d) a transformation of the PNC vowel system which involved delabialization of PNC *o.

The above is based on the article by Starostin (PDF, in Russian) already cited by Octaviano in another thread.

[Octaviano]

My primary purpose in this thread is to discuss my ideas about Indo-Uralic and get some useful feedback, but I'm also very interested other people's ideas.

Ha, ha, ha. You seem more interested in my own theories (to criticise them) than yours.

[Legion]

[Morrígan]

Octaviano should try being more like Basilius.

[Octaviano]

Octaviano should try being more like Basilius.

Please explain.